Embed Size (px)
Citation preview
1
ENGINEERS IRELAND
__________________________________________________________________________________ DEEP EXCAVATIONS IN DUBLIN
RECENT DEVELOPMENTS
_________________________________________
Michael Looby, Director, Byrne Looby Partners.
Dr. Mike Long, Senior Lecturer, University College Dublin.
Paper first presented to a meeting of the Geotechnical Society of Ireland at Engineers Ireland, 22 Clyde Rd, Dublin 4, on 11th December
2007.
Photograph shows Westgate 14 m deep excavation in June 2006
SYNOPSIS
A number of Deep Excavations up to 23m in depth have recently been completed in Dublin. Different approaches including propped and
unpropped, Secant and Contiguous Pile Wall Solutions have been employed on various projects. The paper updates a database for propped
and cantilevered wall supported excavations in Glacial Tills. A comment and interpretation of recorded wall movement versus retained
heights and wall stiffness is provided. Modelled predications are also discussed. A number of case histories of deep basement excavations
including Spencer Dock in the Docklands, 14m excavation at Westgate (Heuston Square) and other projects are presented and discussed.
2
1. INTRODUCTION
The recent period of sustained economic
growth in Ireland has led to an increase in the
use of underground space, with some
development now including 4 underground
levels. This period has also seen the
development of marginal sites, for example
areas of Dublin docklands, which previously
would have been considered unsuitable for
deep basement construction.
The purpose of this paper is to provide an
update on recent developments in deep
excavations in the Dublin area. Specifically
the paper will:
• Briefly review the background geology
• Summarise the situation as existed up to
about 2002
• Present recent developments in the use of
cantilever retaining walls.
• Review the lessons learned from deep
open cuts.
• Present current approaches for the
analysis, design and construction of deep
excavations by reference to some case
histories namely:
→ 7 m cantilever wall at Ballycullen Rd.
where monitoring information for
some 140 weeks is available
→ 14 m excavation at Westgate
supported by a single row of anchors
→ 7 m deep excavation in complex
ground conditions at Spencer Dock in
Dublin docklands.
Finally some issues related specifically to
construction will be discussed and some
recommendations given for future works.
2. BACKGROUND GEOLOGY
Bedrock in the Dublin area is a thin to
medium interbedded homogenous grey
argillaceous limestone and calcareous shale.
Over much of the city, it is overlain by
glacial deposits, known colloquially as
Dublin boulder clay (DBC). This is hard
lodgement till which was deposited beneath
the ice sheet that covered much of Ireland
during the Pleistocene period. It was known
that the ice thickness in Dublin was
approximately 1 km and that several
advances and retreats of the glaciers occurred
in the area. The grinding action of this sheet
as it eroded the underlying rocks coupled
with its loading effect resulted in the
formation of a very dense / hard low
permeability deposit, which contains pockets
of lenses of coarse gravel, particularly at
depth. Oxidiation of the clay particles in the
top 2 m to 3 m has resulted in a change in
colour from black to brown and a lower
strength material.
With the construction of the Dublin Port
Tunnel, a clearer understanding of the
detailed geology of these deposits has
emerged, see Skipper et al. (2005). The
details of the engineering properties and
engineering behaviour of DBC has been
reported by Farrell and Wall (1990), Long
and Menkiti (2007a and 2007b)
Geological conditions in the Dublin
docklands are complex and comprise a series
of estuarine clays, slits, sands and gravels.
The situation in the docklands area is
complicated by the presence of a pre-glacial
channel just north of the River Liffey which
was identified by Farrington (1929). It
diverges from the present channel of the
River Liffey near Connolly Station and
returns near the mouth of the river. It is not
clear whether widely varying sea levels,
tectonic movements or some weakness in the
underlying rock gave rise to the channel.
However from an engineering point of view,
it has significant importance in that is it
generally filled with deposits of glacial and
fluvio-glacial gravels. A study of the deposits
in this area is currently being carried out at
UCD (Research student Brian Kearon).
Useful information can also be found on the
website of the Geological Survey of Ireland
(www.gsi.ie).
Figure 1. Dáil Eireann 6 m excavation
adjacent to (a) Senate chamber and (b)
Trinity College
3. SITUATION UP TO 2002
Up to about five years ago, basements in
Dublin generally comprised two underground
levels and were often constructed within
lightly supported contiguous or secant piled
retaining walls. Some typical examples of
these projects are the 6 m deep excavation at
Dáil Eireann, see Figure 1A and the 7.4m
excavation at Trinity College (see Figure 1B)
(Brangan, 2007)
A possible exception to these early
developments was that of the Jervis St.
shopping centre, where the secant piled
retaining wall had to be designed for an 8 m
deep dig and to support large loads from the
old Jervis St. hospital façade and the adjacent
Marks and Spencer store on Mary St., see
(Dougan et al., 1996).
In general these retaining systems behaved
very well. Lateral wall movements were very
small and prop forces were less than
traditional design approaches would predict
in boulder clay. At Jervis St and the Dáil
some attempts were made to monitor the prop
forces and they were found to be dominated
by temperature effects. Loading resulting
from the excavation was close to zero.
Brangan and Long (2001) (see also Brangan,
2007 and Long 2002b provided a summary
of the situation up to that time and included
data from 9 propped or anchored walls and 3
cantilever walls. All of these sites were
underlain by competent glacial deposits.
3
Figure 2. (a) Mespil Rd (b) Tallaght
Town Centre
The conclusions of this work were:
• The support systems behaved in a very
stiff manner with displacements and prop
forces being much lower than for world
wide stiff soil cases,
• In general traditional wall designs are
conservative,
• In order to properly model the behaviour
soil parameters such as the coefficient of
in situ horizontal stress (K0), the
variation in stiffness with strain and the
undrained shear strength (su) are very
important,
• The pore pressure behaviour during
excavation is poorly understood and it is
likely that high negative pore pressures
(suctions) develop.
Figure 3. Dublin glacial till database
for propped walls (a) δδδδh versus H (b) δδδδh
/ H versus EI/γγγγws4
4.0 UPDATED DATABASE FOR
PROPPED WALLS
The database of Brangan and Long (2001) /
Brangan (2007) and Long (2002b) for
propped walls in competent glacial deposits
is reproduced in Table 1 below and has been
augmented by data from eight other sites
including the 14 m deep Westgate excavation
and data from the Dublin Port Tunnel project
where excavation depths were up to 25 m
(see Figure 4).
A plot of maximum measured lateral
movement (δh) versus retained height (H) is
shown on Figure 3a. Except for the project in
Tallaght (see Figure 2B) all δh values are less
than 10 mm. There does appear to be some
weak tendency for an increase in δh with H.
The stiff behaviour of the very deep Westgate
and Dublin Port Tunnel excavations are
particularly worthy of note.
Also shown on Figure 3a are lines
representing normalised movement (δh/H) of
0.18% and 0.4%. The former relationship
was obtained by Long (2001) for an average
of 169 case histories worldwide where there
was stiff soil at dredge level. The behaviour
of the Dublin projects is significantly stiffer
than the worldwide average. The 0.4% line
represents a typical design value as
recommended by CIRIA report C580 (Gaba
et al., 2003) and clearly this relationship is
very conservative for the Dublin cases.
0 5 10 15 20 25Excavation depth, H (m)
0
10
20
30
40
Ma
x.
late
ral w
all
mo
ve
me
nt,
δh (
mm
)
δh/H = 0.18%
169 world casesLong (2001)
δh/H = 0.4%
typical designGaba et al. (2003)
DPT - WA2DPT
W'gate
DPT
0.1 1 10 100 1000Clough et al. (1989) system stiffness (EI/γws4)
0
0.1
0.2
0.3
0.4
0.5
δh /
H (
%)
δh/H = 0.18%
δh/H = 0.4%
Tallaght
4
Figure 4. Shaft WA2 – Dublin Port
Tunnel Project - 56 m diameter 25m
deep diaphragm wall shaft with single
ring beam support (Photo courtesy Dr.
Chris Menkiti, GCG)
The data shown on Figure 3a takes no
account of the retaining wall type, its
stiffness nor the prop / anchor configuration.
In order to attempt to include these factors,
the data are replotted on Figure 3b in the
normalised form of δh/H against Clough et
al. (1989) system stiffness. This is defined as:
4s
EIstiffnessSystem
wγ=− (1)
where:
EI = wall stiffness
γw = unit weigh of water (required to make
expression unit less)
s = support spacing
Lateral movements appear to be independent
of stiffness. This suggests that a more flexible
(an hence a more economic) wall may
perform adequately in many cases. These
data confirm the findings of the earlier work
that design of retaining wall systems in
Dublin glacial deposits are very conservative.
5. RECENT DEVELOPMENTS IN
CANTILEVER WALLS
5.1 General
Cantilever walls form ideal temporary or
permanent works. The site remains free of
internal props or struts allowing construction
to continue without obstacles. In particular
perimeter works such as drainage can
proceed without hindrance. Difficulties
associated with ground anchorages, such as
obtaining wayleaves or proof testing are also
avoided. Soil mechanics textbooks suggest
cantilever walls are only suitable for modest
excavations up to about 4.5 m retained
height. However throughout the world and
more recently in Ireland cantilever walls have
been used to retain excavations significantly
greater than 4.5 m. In the following sections
the performance of cantilever walls
worldwide and in Ireland will be reviewed
and attempts will be made to explain why the
behaviour of these structures surpasses that
suggested by conventional soil mechanics.
5.2 Cantilever walls worldwide
A summary of the case histories used is given
on Table 2. Note this table is an expanded
version of the one given by Long (2001) and
only those references not quoted in this latter
paper are given here. Figure 5a shows the
maximum measured lateral movement of the
walls (δh) versus retained height (H). The
data have been sub-divided into two groups,
those where there was stiff soil at the bottom
of the excavation (dredge) and where there
was soft soil at dredge. For the stiff soil at
dredge cases, surprisingly, the δh values are
confined to a relatively narrow band certainly
up to a retained height of about 9 m. δh
values are on average about 15 mm. Beyond
H = 9 m, there is a tendency for increasing δh.
For the limited number of cases with soft soil
at dredge level, the movements are greater,
especially for the case of the 10 m deep
excavation in San Francisco.
Data are presented in normalised δh/H versus
H form on Figure 6a. Again the δh/H values
are confined within a relatively narrow band
and seem independent of H, certainly up to H
= 10 m to 11 m. Average δh/H is about
0.25%.
The normalised movement data are
furthermore plotted against Clough et al.
(1989) system stiffness on Figure 7a, in an
attempt to examine the influence of the
retaining system on the movement. Here the
support spacing s is taken to be either the
retained height (H) or H + fixity. Typically s
is taken as 1.4 H. Beyond a Clough et al.
(1989) system stiffness value of 1 (equivalent
to a sheet pile of average size), the lateral
movements appear to be independent of
stiffness. This suggests that a more flexible
(an hence a more economic) wall may
perform adequately in many cases. World-
wide design practice of cantilever walls,
especially where there is stiff soil at dredge
level, may therefore be conservative.
5.3 Cantilever walls in Dublin
A summary of Dublin case histories is also
shown on Table 2. Relatively high cantilever
walls have been used for some time. Use of
these walls was based on the practical
experience of open cuts in the glacial tills
being able to stand unsupported at very steep
angles. Some examples of contiguous piles
retaining walls from Intel in Leixlip and
Charlemont Place, in Ranelagh, Dublin are
shown on Figure 8 (Long, 1997).
More recently cantilever walls of 7.5 m or so
are being used regularly. An example of the
7.5 m high cantilever 600 mm diameter
contiguous pile wall at Hunters Wood,
Ballycullen Rd. is shown on Figure 9. The
design, construction and performance of this
structure will be discussed in more detail in a
later section of the paper.
56m diameter Diaphragm Wall TBM Launch Shaft56m diameter Diaphragm Wall TBM Launch Shaft
5
Figure 5. Cantilever retaining walls δδδδh
versus H (a) world-wide case histories,
(b) Dublin
Figure 6. Cantilever retaining walls
δδδδh/H versus H (a) world-wide case
histories, (b) Dublin
Figure 7. Cantilever retaining walls
δδδδh/H versus Clough et al. (1989) system
stiffness (a) world-wide case histories,
(b) Dublin
2 4 6 8 10 12Excavation depth, H (m)
0
50
100
150
200
250
Ma
x.
late
ral m
ove
., δ
h (
mm
)
Stiff soil at dredge
Soft soil at dredge
World-wide case histories
2 4 6 8 10 12Excavation depth, H (m)
0
50
100
150
200
250
Ma
x.
late
ral m
ove
., δ
h (
mm
)
Stiff soil at dredge
Soft soil at dredge
Dublin case histories
Trendline from world-widecase histories
2 4 6 8 10 12Excavation depth, H (m)
0
0.5
1
1.5
2
2.5
Norm
alis
ed m
ax. la
tera
l m
ove
., δ
h / H
Stiff soil at dredge
Soft soil at dredge
World-wide case histories
2 4 6 8 10 12Excavation depth, H (m)
0
0.5
1
1.5
2
2.5
No
rmals
ied m
ax. la
tera
l m
ove.,
δh
/ H
Stiff soil at dredge
Soft soil at dredge
Dublin case histories
Trendline from world-widecase histories
0.01 0.1 1 10 100 1000
Clough et al. (1989) system stiffness = EI/γwh4
0
0.5
1
1.5
2
2.5
Norm
alis
ed m
ax.
late
ral m
ove
., δ
h / H
Stiff soil at dredge
Soft soil at dredge
World-wide case histories
0.01 0.1 1 10 100 1000
Clough et al. (1989) system stiffness = EI/γwh4
0
0.5
1
1.5
2
2.5
No
rma
lsie
d m
ax. la
tera
l m
ove
., δ
h /
H
Stiff soil at dredge
Soft soil at dredge
Dublin case histories
Trendline from world-widecase histories
6
Figure 8. Cantilever retaining walls at
Intel, Leixlip (6.8 m max.) and (b)
North Wall Quay (7.5 m max.) Sept
2007
Figure 9. 7.5 m high cantilever wall at
Ballycullen Rd. (a) during construction
in June 2005 and (b) in service
September 2007
Plots of δh versus H, δh/H versus H and δh/H
versus Clough et al. (1989) system stiffness
for the Dublin case histories are shown on
Figures 5b, 6b and 7b respectively. In each
case the best-fit trendline from the worldwide
database of cantilever walls are also shown.
The Dublin walls have performed very well,
with values falling in general well below the
trendlines. Average δh and δh/H values are
about 5 mm and 0.08% respectively. An
exception is the Thorncastle St. case history
(Long et al., 2002) where there was soft
alluvial soil at dredge level.
Again it seems that lateral movement is
independent of system stiffness and once a
sufficiently stiff system is provided the walls
will behave well. Again it seems Irish
practice is conservative and perhaps even
more conservative than that worldwide.
Based on these data it seems there is scope
for the use of higher cantilever walls, at least
for temporary work s purposes.
5.4 Behaviour with time
The data presented above omit one very
important factor, i.e. how does the lateral
movement vary with time. This has important
implications as to whether these walls can be
used for permanent work as well as
temporary works and also in the temporary
case how long is the useful life span.
Data from five sites in Dublin, Ballycullen
Rd., Cork St., New St. and the Dublin Port
Tunnel all in glacial till and Thorncastle St.
in soft alluvial soils are shown on Figure 10a.
The data for Ballycullen Rd is of particular
interest as it spans a period of some 140
weeks (2.7 years). In most of the projects,
after a relatively short time, the retaining wall
was encorporated into the permanent works
meaning it was no longer acting as a
cantilever. The Ballycullen Rd site is unusual
as the wall forms the permanent works, see
Figure 9b. On Figure 10b the data for the
first 40 weeks is shown in more detail.
It can be seen that in all cases there is a
gradual development of movement increasing
with time. For the glacial till cases the rate of
increase of lateral displacement is relatively
slow. However for the soft alluvial soils case
at Thorncastle St the development of
movement is rapid.
7
Figure 10. Cantilever retaining wall
movement with time
The reason for this behaviour is the gradual
dissipation of negative pore pressures
(suctions) and the build up of positive pore
pressure. This will be discussed in the
following sections.
6. LESSONS LEARNED FROM
STEEP UNSUPPORTED CUTS IN
DBC
6. 1 Natural slopes
There are several examples of natural steep
slopes in glacial tills in the Dublin area.
Hanrahan (1977) and Long et al. (2003)
describe near vertical slopes of up to 33 m
high in Howth (north of Dublin), Killiney
(south), Greystones (south of Dublin in Co.
Wicklow) and Chapelizod (west). There are
similar high natural cuts along the River
Dodder, in the south of the city
Figure 11(a) Glacial till sea cliffs at
Greystones, Co. Wicklow, (b) man made
6 m cut in Rathfarnham, south Dublin.
A view of the cliffs at Greystones is shown
on Figure 11a. Local instability can be seen,
for example when the slope was undercut by
sea erosion.
6.2 Man made temporary cuts
Temporary cuts of significant thickness,
made for engineering works, can stand
unsupported for relatively long periods, see
Figure 11b for example. Groundwater level
(as recorded by piezometers) was at about 2
m depth. The cut was made during the
summer time and is mostly in the weathered
upper brown boulder clay (UBrBC). No
ground water control system was used within
the excavation and no cracks were observed
at the slope crests, for a period of several
months. The authors have seen other near
vertical 8 m deep cuts in central Dublin stand
unsupported, adjacent to heavily trafficked
roads, for periods of several months.
6.3 Instability in steep cuts
Some instances of instability of cuts in
Dublin boulder clay have been reported. For
example Hanrahan (1977) discusses failures
caused by water bearing granular layers in
slopes in Glencullen, south Dublin. During
December 2000, the railway line adjacent to
the sea cliffs at Killiney had to be closed on
three occasions due to landslides. These were
due to surface slips in the upper (weathered)
portion of the till which occurred after
periods of high rainfall. Some local
instability can be seen in the slope beneath
the Martello tower at Howth, Co. Dublin.
This failed in December 2002 following a
period of heavy rain and caused the road
underneath to be blocked.
Long et al. (2003) describe some minor
failures in an 8 m high steep open cut in
north Dublin.
0 40 80 120 160Time (weeks)
0
5
10
15
20
25
Maxim
um
late
ral w
all
dis
pla
cem
en
t (m
m)
0 10 20 30 40Time (weeks)
0
5
10
15
20
25
Thorncastle 2
Thorncastle 4
Ballycullen 1
Ballycullen 2
Cork St. 1
Cork St. 2
New St. 3
New St. 5
DPT - DP79
8
Menkiti et al. (2004) describe the failure at
the trial trench excavated for the Dublin Port
Tunnel project.
The failures typically occur after periods of
high rainfall, at the location of granular
lenses and are characterized by the slip of
thin flakey wedges.
The behaviour of these natural and man made
cuts should give some clues as to the
surprisingly stiff behaviour of the Dublin
retaining walls.
6.4 Possible reasons for stable steep cuts
The possible reasons why steep cuts in
Dublin glacial till can stand unsupported for
significant time periods are discussed in
detail by Long et al. (2003). Possible
explanations for this observed behaviour are:
• The till possesses a high drained
cohesion (c') component.
• The particles are cemented together.
• Pore water suctions.
Long and Menkiti (2007a and 2007b) have
explained why the depositional nature of the
till and its particle size distribution have
resulted in it having a curved rather than the
normally assumed linear failure surface with
c' = 0. Evidence from scanning electron
microscope photographs has confirmed no
appreciable cementation exits between the
soil particles. Therefore pore water suction
must perform a significant role in the
observed behaviour of the material. This is
consistent with the failure of the material in
the presence of granular lenses and high
rainfall.
Figure 12. Pore water pressure
reduction in cut slopes
6.5 Pore water suctions
The pore water pressure conditions in the
slope before and after excavation are shown
diagrammatically on Figure 12. Before
excavation the pore water pressure (u) is
simply given by the hydrostatic head below
the groundwater table, which is at about 2 m
depth, i.e.
zu wγ= (2)
The change in u caused by the excavation can
be determined (approximately at least) from
Skempton’s (1954) classical equation for
pore water pressure change:
( )[ ]313 σσσ ∆−∆+∆=∆ ABu (3)
In this case both the all round pressure (∆σ3)
and the deviator stress (∆σ1-∆σ3) reduce due
to the excavation-induced stress relief. This
means that u reduces and, depending on its
initial value, could become negative. If u
reduces then the effective stress increases,
thus improving, in the short term, the
stability of the slope. The length of time over
which this reduced pore water pressure can
be sustained is a complex issue and depends
on the soil type, its fabric, permeability, the
sequence of construction, slope protection,
weather, etc.
As well as increasing the strength of the soil,
from the point of view of retaining wall
design the net effect is that there will be none
or at least very low pore water pressure
acting on the wall. Normally designers
assume hydrostatic water pressures, which
can be higher than soil pressures.
As can be seen from Skempton’s formula
above it is necessary to determine ∆σ3 and
∆σ1-∆σ3 in order to calculate the pore water
pressure after excavation. This is not a trivial
matter. Some elastic based solutions exist. In
the Dublin Port Tunnel project use was made
of the finite element approach. Further details
of these analyses can be found in Menkiti et
al. (2004) and Kovacevic et al. (2007). This
work was carried out by the Geotechnical
Consulting Group (GCG), who made use of
the Imperial College, London, geotechnical
finite element code called ICFEP. Some
typical output for the Dublin Port Tunnel
trial trench excavation is shown on Figure
13.
6.6 Some examples of measured pore water
suctions
The techniques used for measuring pore
water suctions on the Dublin Port Tunnel
project are described in detail by Long et al.
(2004). This includes a description of the
special piezometers used and the method
used to seal them into the ground.
A typical example for a piezometer at about
5.5 m depth in the trial trench is shown on
Figure13b. It can be seen that the measured
suction of about –30 kPa is very similar to
that predicted by the finite element analysis.
To the authors knowledge there has been no
measurements made of pore water suctions
behind retaining walls in Dublin boulder
clay. Such data would be of significant
practical use in the future design of these
systems.
12 mfor DPT
Groundwater table at 2 m
Before excavation pore pressure, u = γw x z
After excavationstress relief reduces
u and it possibly becomes negative (i.e suction)
z
Slope angle 70 to 80 deg. for DPT
Before excavation After excavation
9
Figure 13a. Predicted pore water
pressures for Dublin Port Tunnel trial
trench and (b) typical measured value
for piezometer at 5.5 m (Long et al.,
2004)
7. OTHER SOIL FACTORS
INFLUENCING RETAINING WALL
BEHAVIOUR
In addition to the role of pore water suctions
there are other significant factors influencing
the very stiff behaviour of retaining walls in
Dublin boulder clay. These are best
illustrated by examining a series of high
quality isotropically consolidated (CIUC)
triaxial tests (carried out by NMTL Ltd. for
ARUP Consulting Engineers) for a site in the
south city centre. The tests were carried out
on GeoBore-S rotary cores.
From the test results it can be seen:
• The material is very strong with
undrained shear strength (su) typically
between 400 kPa and 500 kPa
• It is ductile with peak strength being
developed at relatively high strains. (This
is advantageous from the point of view of
steep slopes or retaining wall moments as
the signs of incipient failure will be seen
some time before failure occurs).
• It is highly dilatant with negative pore
pressures developing during shear.
• Stiffness is highly non linear.
• Stiffness values are very high. Shear
modulus (G, as measured by local sample
mounted transducers) varies between 400
to 600 MPa at strains of 5x10-4 % to less
than 100 MPa at 0.01%.
• Stiffness measured by local strain
transducers in the laboratory is consistent
with very small strain stiffness measured
in situ using MASW surface wave
techniques (Donohue et al., 2003)
Figure 14. CIUC triaxial test
results for Dublin boulder clay
(Courtesy NMTL Ltd. And
ARUP Consulting Engineers)
Scale10m0
-75kPa-50kPa
-25kPa 0kPa
125kPa
100kPa
75kPa
50kPa
25kPa
-40.00
-30.00
-20.00
-10.00
0.00
10.00
20.00
30.00
40.00
2/23/02 3/5/02 3/15/02 3/25/02 4/4/02 4/14/02 4/24/02
Po
re P
res
su
re (
kP
a)
Piezometer 3B - 5.5 m - Trial trench
Excavation
10
8.0 DESIGN APPROACH
The previous sections have shown that low
level displacement has occurred for a number
of various types of retaining walls, primarily
in boulder clay soils. From a practitioners
point of view in order to understand why low
level deflections are occurring it is important
we understand the background to retaining
wall design. The following section briefly
outlines the general pile design approach and
the issues critical to wall movement.
As with any structure the design approach for
basement retaining structures commences
with the overall design philosophy and
proceeds through to detailed numerical
design. Assuming that the site investigation
has been completed, the initial considerations
will include the following issues which are
separated into those critical to movement and
other issues:
Movement Related Issues:
• Allowable movements
• Ground Model - Cohesive or
Cohesionless
• Quality of Site Investigation
• Depth of required excavation
• Prevailing groundwater conditions
• Retained structures and services –
surcharges
• Propped/cantilevered
• Temporary/Permanent retention
Other Issues:
• Space limitations/available plant
• Budget and programme
• Vertical load carrying requirement
• Recovery of material (e.g. sheet
piles)
• Working space (single vs. double
sided shutters for permanent RC
walls)
Once the broad requirements, based on the
above criteria, have been established then the
design can proceed to the numerical design
stage.
8.1 Limit States
Simpson and Driscoll (1998) define limit
state design as a procedure in which attention
is concentrated on avoidance of limit states,
i.e. states beyond which the retaining wall no
longer satisfies the design performance
requirements. This relates to the possibility
of damage, economic loss or unsafe
situations.
Retaining walls should be designed for
Ultimate Limit States (ULS) and
Serviceability Limit States (SLS).
8.2 Ultimate Limit States
Ultimate limit states are those associated with
collapse or with other similar forms of
structural failure. They are concerned with
the safety of people and the safety of the
structure.
The following should be considered:
• Loss of equilibrium of the structure
• Failure by rotation or translation of
the wall
• Failure by lack of vertical
equilibrium of the wall
• Failure of a structural element such as
wall, anchor, prop or wailing beam
• Movements of the retaining structure
that may cause collapse of the
structure, nearby structures or
services which rely upon it
• Failure caused by fatigue or other
time-dependent effects.
The ULS design involves carrying out a
Limit Equilibrium Analysis. Appropriate
Factors of Safety are applied and earth
pressure diagrams are derived, using classical
earth pressure theory in which full “Active”
and “Passive” conditions are assumed.
Computer programmes allow rapid analyses
to be carried out and readily allow for
changes to geometry, stratigraphy, soil
parameters etc.
This type of analysis will produce a set of
bending moments and shear forces in the wall
along with calculated propping forces. These
forces may or may not be the actual design
forces and hence cannot be used to predict
deflection.
8.3 Serviceability Limit States
Serviceability limit states correspond to
conditions beyond which specific service
performance requirements are no longer met,
for example pre-defined limits on the
amounts of wall deflections.
The permissible movements specified in the
design should take into account the tolerance
of nearby structures and services to
displacement.
It is worth remembering that the overall
stability analysis was carried out using full
“Active” and “Passive” soil conditions with
factored soil and surcharge parameters.
The next stage of the analysis procedure is to
calculate the actual wall, ground and building
movements along with the forces in the wall
in the “serviceability condition”. Soil
generally does not exist in the ground, pre-
development, at either it’s active or passive
limit and generally will not fully reach its
passive limit during the works, as appropriate
factors of safety will have been employed.
Importantly, the soil may not reach its active
limit either and therefore the soil load on the
wall in the SLS condition may be greater
than in the ULS condition.
As stated above bending moments, shear
forces and prop loads were calculated based
on the soil model presented in the ULS
analysis. However if the soil does not fully
reach its passive (or active) limit then this
model will not be accurate and an alternative
model or models are required.
Before construction activities occur soil
exists in-situ at “At-rest” or “Ko” (co-
efficient of earth pressure at rest) conditions,
see Figure 15. As soil cannot strain laterally
as it is being compressed during its
formation, lateral stresses are generated. This
subject has generated a vast amount of
research but for the purpose of this paper a
simple example will be used.
Figure 15 Earth Pressure at Rest Ko
As an excavation proceeds, see Figures 16,
the earth pressures move from the Ko
condition towards either the Ka or Kp
condition. The magnitude and distribution of
the shift and the “serviceability” pressures on
the wall are based on numerous factors,
however the stiffness of the various soil strata
have a significant influence.
Figure 7 Notional SLS Earth Pressures
Figure 16 – Ko Condition
It is therefore apparent that the SLS pressures
are different in magnitude and shape to those
in the ULS model. Therefore the ULS model
cannot be used to accurately predict the
displacement of the wall.
11
Modern SLS analyses are normally carried
out using finite element (FE) or finite
difference (FD) techniques and there are
many computer packages available to carry
out these. An example from the PLAXIS
package is presented in Figure 17 below.
Figure 17 - Example of PLAXIS
Output
The serviceability analyses are carried out
using unfactored soil parameters with no
over-dig allowance.
8.4 Modelling Method/Approach
After horizontal or vertical unloading
theoretical soil and groundwater pressures in
cohesive soils can be significantly less than
zero and can sometimes provide no active
load on a retaining wall. This will be further
discussed below. For a number of reasons
this is not an acceptable design approach as
set out below:
• Tension cracks can develop to the
rear of the wall. The relevant code
(CIRIA 580) requires the use of a
Minimum Equivalent Fluid
Pressure (MEFP) for tension cracks
• Sand and gravel lenses which are a
common occurrence in the Dublin
boulder clay can supply sufficient
groundwater to allow hydrostatic
groundwater pressures develop to
the rear of the wall
• Potential unidentified pockets of
sand and gravel could provide
loading conditions and failure
modes that tend towards effective
stress conditions. (This condition
applies to the passive side also).
In order to account for the above criteria in
retaining wall design BLP currently use one
of the following approaches (a key factor in
which approach is employed is the level of
the groundwater for the particular project).
• Effective stress conditions with a
low Ka value
• Undrained soil parameters with a
minimum equivalent fluid pressure
employed in the analysis as per
CIRIA C580.
• Undrained soil parameters with full
hydrostatic groundwater pressures
• A combination of the above criteria
The above paragraphs outline conditions that
can occur and the relevant design criteria. In
reality, full hydrostatic water pressures,
MEFF conditions or effective stress
conditions may not exist and actual loading
conditions on the wall could be closer to the
undrained or partially undrained conditions
where complex (incl. negative) pore
pressures exist. This results in low or zero
lateral pressures on the wall. Obviously the
deflection predictions from the analysis with
undrained parameters will be significantly
less than any of the design approaches
discussed above.
The use of undrained parameters in
conjunction with the observational approach
may be considered for reducing predicted
deflections to simplify the construction
sequence and reduce costs. (In addition to
this, how we model groundwater with time
post excavation should also be considered
but is not addressed as part of this paper).
This approach should also only be considered
where the predicted deflections using
traditional approaches are within defect limits
to prevent the possibility of damage,
economic loss or unsafe situations. The risk
associated with the decision should be clearly
assessed in terms of understanding of the site
geology, type and condition of structures to
the rear of the pile wall and the quality of
monitoring procedures put in place. This
decision to use undrained parameters will
have a large impact on predicted deflections
as discussed above.
The benefits of using undrained parameters
are not so much that pile sizes will be
reduced but that more cost effective overall
solutions can be employed in more areas. It
seems there is scope for the greater use of
cantilever walls and also greater retained
heights at least for temporary works
purposes.
Similar design issues arise in relation to
undrained soil parameters on the passive side
of the wall and undrained analysis can
potentially allow for very shallow
embedment depths. In order to allow for the
possibility of unknown gravel layers,
effective stress parameters are often used to
determine the overall stability requirements.
However the serviceability analysis may
require the benefit of using undrained
material on the passive side in order to more
accurately predict deflection.
8.5 Quality of Site Investigation Data
As with all geotechnical applications, quality
of site investigation is critical to accurately
predict movements. The results of recent
consolidated triaxial tests presented in
Section 7.0 are consistent with monitored
pile deflections in boulder clay, in that they
suggest that we are currently significantly
under estimating the stiffness parameters of
the glacial tills. Higher quality laboratory
testing on larger projects should be
considered.
Investigations should accurately establish the
static groundwater profile at the site. As
noted previously, to the author’s knowledge
there have been no measurements made of
pore water suctions behind retaining walls in
Dublin boulder clay. Such data would be of
significant practical use in the future design
of these systems.
8.6 Critical Issues for Movement
A summary of the critical issues in relation to
retaining wall deflections and predictions are
as follows:
• Cohesive or Cohessionless model –
in conjunction with how we model
groundwater with time
• Continuous Wall (Secant/sheet pile)
or contiguous
• Surcharge
• Propped/Cantilevered
• Groundwater conditions
• Soil parameters – quality of
investigation
• Attitude to risk
9.0 CASE HISTORIES
The design and monitoring of 3 No. deep
basements are described in the following
sections. All projects were completed in the
past three years. The projects include
Spencer Dock Development at North Wall
Quay Wall, the Westgate (now called
Heuston Square) project and a residential
development in Dublin 24.
12
Figures 18/19 - Spencer Dock and
Westgate
The projects were selected to compare the
performance of permanent and temporary,
propped and cantilevered walls in various
soils types and the accuracy of predicted
deflections.
9.1 Westgate
The Westgate Development (see Figure 19) is
a combined commercial and residential
development, adjacent to Heuston Station,
south of the Liffey. Topographical levels on
the site vary between 14m OD at the south
end of the site and 6m OD at the north.
Basement formation was 0m OD resulting in
a 14m retained height at the southern end.
Ground stratigraphy consisted of stiff to hard
glacial till interlayered with dense gravel.
The gravel layers were intermittent and did
not exist consistently across the site. A
geological x–section of the site is provided in
Figure 20. Soil parameters employed for the
different strata are as indicated in Table 3.
A monitoring programme indicated that
groundwater levels varied between +8m OD
of the southern boundary and 0.9m OD at the
northern boundary. The variable levels were
due to groundwater flow towards the River
Liffey.
Construction Sequence
The construction sequence was as follows:
• Install 900mm φ “Secant” Soft-Hard
Pile wall to 20m bgl at 0.7m cc to
achieve overlap. Piles were
constructed with a 30 Tonne metre
Torque CFA Rig. The male piles were
•
•
• `
reinforced with 18 No. T32’s over the
central third of the pile and 10 No.
T32’s over the top and bottom third of
the pile and were constructed with
• C35 Concrete.
• Construct a reinforced capping beam.
• Install, test and pre-stress 700 kN SWL
anchors @ 2.25m cc through the
capping
• Complete excavation in staged manner
with a detailed monitoring programme.
Table 3 - Westgate Soil Parameters
Figure 21 illustrates a typical x-section
through the pile wall.
A key aspect of the retaining wall design
was the use of the “Observational
Approach” to monitor the movement of the
wall as excavation proceeded. 6 No.
inclinometers were installed at key locations
across the site to measure pile wall
deflection. Inclinometer monitoring was
carried out by BLP in-house with
monitoring being carried out at critical
stages of the excavation.
“Trigger level” lateral movement criteria was
established prior to the works commencing
for the different stages of excavation. The
original design analysis predicted pile
deflections of 50mm, the pile was structurally
designed to resist the bending moment that
reflected this profile. The maximum trigger
level value of 35mm for complete excavation
was selected to ensure that excessive
deflections did not have an impact on the
existing listed boundary wall to the rear of
the pile wall.
The following protocol was agreed if the
monitored deflection exceeded the trigger
value.
1. Excavation works would cease.
2. A second set of readings would be
taken the following day.
3. If further movements were
observed the wall would be
partially backfilled.
4. If the wall had stabilised, a second
layer of the tie-back anchors
would be installed.
Recorded Movement
Figure 21 illustrates a typical x-section
showing the original predicted deflection and
the recorded pile inclinometer data. The
original analysis included modelling the soil
stratigraphy with drained parameters using
both FREW and PLAXIS analyses. In the
PLAXIS analyses the made ground and
glacial soils were assumed to behave as
elastic perfectly plastic materials with failure
defined by Mohr Coulomb using drained
parameters. The analyses gave reasonably
consistent predictions within 20% of each
other.
Maximum recorded pile deflections were
12mm. The modelled movements are
significantly different. It is noted also that
while the original Site Investigation
suggested that significant gravel layers may
be present, inspection of the bulk excavation
at the southern end of the site indicated the
soil was predominantly a clay material.
Stratum Depth
m bgl
E
MPa
Ko φφφφ (o) Cu C
Fill 0- 1 50 .5 30 - -
Dense
Gravel
1-25 150 1 38 - -
Stiff
Brown
Clay
1-25 150 1 35 - -
13
Figure 20 Westgate Geological Section
For the purposes of this paper an analysis of
the wall with undrained soil parameters and a
static water level 7m OD was carried out
using FREW. Predicted pile deflections were
20mm suggesting that the undrained
parameters would model actual pile
deflections more accurately. Figure 21 also
indicates the predicted pile deflection profile
using the drained parameters. The
introduction of a consolidation stage in the
analysis to model groundwater pressures with
time may further improve predictions.
9.2 Spencer Dock
Spencer Dock (see Figure 18) is located in
the Dublin Docklands and is 2km east of the
city centre adjacent to the River Liffey.
Topographical level at the site is +2.5m OD.
The site is bounded to the west by the
Spencer Dock canal and the south by the
River Liffey.
Figure 22 illustrates the typical geology of
the site.
The water table on the site is 0m OD and is
not significantly influenced by the tidal cycle.
The upper made ground consists of brick and
masonry in a clay matrix. Due to the variable
depositional environment in this section of
the site close to the river, the alluvium
stratum is a complex mixture of soil types
typical of the docklands. The upper 3.5m
comprises a loose to medium dense sand and
gravel. This is underlain by a soft clay and
silt with organic material.
N-values in the soft clay layer vary from 2 to
7. The underlying glacial deposits consist of
2 to 3m of dense gravel overlying a hard till.
Construction Sequence
A 900mm φ soft/hard secant pile wall was
employed. Tie back anchors were employed
at approximately 2.5m bgl rather than at
capping beam level. The construction
sequence therefore involved an initial
cantilevered stage of 2.5m.
Piles were extended to 16m bgl to include a
2m embedment into boulder clay to achieve
groundwater cut-off. As with the Westgate
project a detailed inclinometer monitoring
programme was implemented to monitor pile
movement.
Movement
FREW was employed to carry out the
original pile wall analysis and indicated that
predicted pile deflections would be in the
range of 35mm to 40mm. A sensitivity
analysis was carried out at the time of the
design varying the stiffness and strength
characteristics of the soils as well as
modelling the layer with drained and
undrained parameters for the soft alluvium
layer.
Inclinometer readings over the course of the
construction stage recorded the deflection
profile as being reasonably consistent with
modelled deflections, approximately 15%
less as per Figure 22, illustrating that the
model may be better at predicting movements
in the cohesionless soil layers and low
strength strata’s as encountered.
9.3 Ballycullen Road
The Ballycullen Road project was a
residential development located in the
Knocklyon area of Dublin 24 (see Figure 23).
The development consists of a number of
apartment blocks constructed over a double
level basement. Datum level at the site was
Figure 21 Westgate Pile Section.
107 mAD and basement formation was
99.5mAD providing a 7.5m effective retained
height. The wall was installed in January
2005 and excavation was carried out in
February 2005.
Groundwater monitoring at the site indicated
that static water levels were below excavation
level.
The basement directly bounds a public access
road to the residential estate necessitating the
requirement for the retaining wall.
The geology of the site consisted
predominantly of glacial till. A brown Firm
to Stiff boulder clay was noted to overlay
Hard Black Boulder Clay as per Figure 23.
The brown and black boulder clay layers
were modelled with effective stress
parameters of 35 and 37 degrees respectively.
The deeper boulder clay layers were
modelled with stiffness parameters of
120,000kPa and 80,000kPa in the temporary
and permanent conditions respectively.
Construction Sequence
The pile wall was required to provide both
permanent and temporary retention (see
Figure 9a & 9b). A 600mm diameter
contiguous pile wall solution was chosen
based on the geology of the site and the
retained height. The construction sequence
was as follows:
• Install 600mm diameter hard piles at
750mm cc to a total depth of 14m below
ground level. Piles were constructed
with a 24 Tonne metre Torque CFA Rig.
The piles were reinforced with 8 No.
T32 over the full height of the pile wall.
14
Figure 22 Spencer Dock Pile Section
• Construct the reinforced capping beam.
• Complete excavation in a staged manner
with detailed monitoring.
• Install a basement concrete slab to
provide a low level prop/restraint.
Prior to construction commencing a number
of design criteria were established including
the following:
• If local gravel layers were encountered
they would be shotcreted to prevent loss
of ground between piles. If more
significant gravel layers were
encountered a Secant Pile Wall solution
would be installed.
• If movement levels in the construction
sequence exceeded a trigger level of
25mm, tie-back anchors would be
installed. The original design analysis
employed drained conditions to the rear
of the wall in the temporary stage.
Predicted pile deflections were 40mm
and 45mm respectively in the temporary
and permanent conditions. The trigger
values were selected to ensure that long
term deflections would not be excessive
in order to avoid damage to the adjacent
road.
Recorded Movement
2 No. inclinometers were installed at key
locations on the pile wall for measuring pile
wall deflections. These have been left in
place to allow monitoring on an ongoing
basis to be carried out and this has been done
over the past 2 years.
Recorded pile deflections during the
construction stage and post construction to
November 2007 (22 Months) were 10mm
and 17mm respectively. Recorded
deflections to date are significantly less that
predicted the deflections of 45mm. An
analysis with undrained parameters was
carried out for this paper which indicated
deflections in the temporary condition of
17mm which is in line with those recorded to
date.
10. CONCLUSIONS
1. Case history data confirms retaining wall
behaviour in Dublin glacial till is
extremely stiff. This applies to
excavations up to 25 m deep.
2. It appears that current approaches over
predict walls deflections and the use of
overall current design practice is clearly
conservative.
3. The use of undrained parameters in
conjunction with the observational
approach may be considered for reducing
predicted deflections to simplify the
construction sequence and reduce costs.
4. This approach should also only be
considered where the predicted
deflections using traditional approaches
are within defect limits to prevent the
possibility of damage, economic loss or
unsafe situations.
5. Cantilever walls have been successfully
constructed up to 7.5 m high. These walls
show smaller movements than expected,
though the development of movement
with time is very important.
6. It seems there is scope for the greater use
of cantilever walls and possibly also
higher retained heights, at least for
temporary works purposes.
7. Important insights into the above can be
gained from observations of steep slopes
in Dublin boulder clay, where pore water
suction plays an important role.
Figure 23 Ballycullen Rd Pile Section
8. Laboratory testing on high quality
samples of the material confirm that it is
stronger and stiffer than normally
assumed in design.
9. Measurement of Pore Water Pressure to
rear of walls should be carried out.
10. The analysis employed accurately
predicted deflections in the boulder clay.
ACKNOWLEDGMENTS
Dr. Chris Menkiti and Dr. George Milligan
for input from DPT project, Dr. Carl
Brangan, former PhD student at UCD. Mr.
Tony O’Dowd, PJ Edwards & Co., Mr. Pat
Fox, Murphy International Ltd and Mr.
Douglas Cook, FK Lowry Piling Contractors.
REFERENCES
1. Bentler, J.G. and Labuz, J.F. (2006).
“Performance of a cantilever retaining
wall”. ASCE Journal of Geotechnical and
Geoenvironmental Engineering, Vol.
132, No. 8, August, pp 1062 - 224.
2. Brangan, C. and Long, M. (2001).
“Behaviour of buildings adjacent to deep
excavations in Dublin glacial till”. Proc.
Int. Conf. Response of Buildings to
Excavation Induced Ground Movements,
Imperial College, London, 17 – 18 July
2001, CIRIA Special Publication No.
199, pp 209 – 216
3. Brangan, C. (2007). “Retaining walls in
Dublin boulder clay” PhD thesis,
University College Dublin, November.
4. Byrne, J. (2007). “Deep Basement
Design and Construction”, The
Institution of Structural Engineers in
association with The Dept. of Civil &
15
Structural Engineering, Dublin Institute
of Technology.
5. Cabarkapa, Z., Milligan, G.W.E, Menkiti,
C.O., Murphy J. and Potts D. M. (2003).
“Design and performance of a large
diameter shaft in Dublin Boulder Clay”.
British Geotechnical Association
International Conference on Foundations,
Dundee, Thomas Telford: 175 – 185.
6. Clough, G.W., Smith, E.M. and
Sweeney, B.P. (1989). “Movement
control of excavation support systems by
iterative design”. Proc ASCE Foundation
Eng: Current Principles and Practices;
2; pp 869-884.
7. Curtis, P. and Doran, J. (2003).
“Retaining wall behaviour at the Dublin
Port Tunnel”. BE Project, UCD.
8. Donohue, S., Gavin, K, Long, M. and
O’Connor, P. (2003). “Gmax from
multichannel analysis of surface waves
for Dublin boulder clay”. Proc. 13th.
ECSMGE, Prague, Vanicek et al., Eds.,
Vol. 2, pp 515 – 520. Published by
CGtS, Prague.
9. Dougan, I, Long M.M. and Byrne J.J.B.
(1996). “The Geotechnical Aspects of the
Deep Basement for the Jervis St.
Shopping Centre”. Trans. Institution of
Engineers of Ireland, Vol. 120, 1996 /
1997, pp 49 - 70.
10. Farrell, E. R. and Wall, D. (1990). “The
soils of Dublin”. Transactions of the
Institution of Engineers of Ireland, 115:
78-97.
11. Farrington, A. (1929). The pre-glacial
topography of the Liffey basin. Proc.
Royal Irish Academy,1929, 38, B, 9. 12. Gaba, A.R., Simpson, B., Powrie, W. and
Beadman, D.R. (2003). “Embedded
retaining walls – Guidance for economic
design”. CIRIA report C580.
13. Hanrahan, E.T. (1977). “Irish glacial till -
origin and characteristics”. RC 164 An
Foras Forbartha, Dublin, 1977.
14. Kovacevic, N., Milligan, G.W.E., Long,
M., and Potts, D.M. (2007). "Finite
element analyses of steep man-made cuts
in Dublin Boulder Clay". Accepted for
publication in Canadian Geotechnical
Journal, July 2007.
15. Long, MM (1997). “Design and
Construction of Deep Basements in
Dublin, Ireland” Proc XIV Int. Conf. on
Soil Mech. and Foundation Engineering,
Hamburg, Germany , September 1997,
Vol. 2, pp 1377-1380.
16. Long, M (2001). “A database for
retaining wall and ground movements
due to deep excavations”. ASCE Journal
of Geotechnical and Geoenvironmental
Engineering, Vol. 127, No. 3, March, pp
203 - 224. See also closure to discussion
on this paper by M. Long in Vol. 128,
No. 6, pp 536 – 537.
17. Long, M., Brangan, C. and Gavin, K.
(2002). “Behaviour of cantilever
retaining walls”. Geotechnical Aspects of
Underground Construction in Soft
Ground (IS-Toulouse 2002), Toulouse,
France, October. Published by
Spécifique, Lyon, ISBN 2-9510416-3-2,
pp 471 – 476.
18. Long, M. (2002a). “Observations of
ground and structure movements during
site re-development in Dublin”.
Institution of Civil Engineers Journal of
Geotechnical Engineering, Vol. 155, No.
4, October, pp 229 – 242. See Also
Discussion in Vol. 156, No. 3, p165.
Published by Thomas Telford Ltd.
19. Long, M. (2002b). “Review of behaviour
of basement excavations in Dublin glacial
till”. Proc. Workshop on Advanced
Laboratory Testing of Geomaterials,
NTU Singapore, Dec., pp 1 –22.
Published by Nanyang Technical
University. Also presented to IEI
Seminar on Basement Design and
Construction, 14 Nov.
20. Long, M., Menkiti, C. O., Kovacevic, N,
Milligan, G. W. E., Coulet, D. and Potts,
D. M. (2003). “An observational
approach to the design of steep sided
excavations in Dublin glacial till”.
Underground Construction 2003, London
UK , September, Published by
Hemmings, pp 443 – 454.
21. Long, M., Menkiti, C.O. and Follet, B.
(2004). “Some experience in measuring
pore water suctions in Dublin glacial
till”. Geotechnical News / Geotechnical
Instrumentation News (GIN), Vol. 22,
No. 3, Sept. 2004, pp 21 – 27. 22. Long, M. and Menkiti, C.O. (2007a).
“Geotechnical properties of Dublin
boulder clay”. Géotechnique, Vol 57, No,
7, September, 595 - 611.
23. Long, M. and Menkiti, C.O. (2007b).
Characterisation and engineering
properties of Dublin Boulder clay.
Proceedings 2nd International Workshop
on Characterisation and Engineering
Properties of Natural Soils (“Natural
Soils 2006”). NUS Singapore, November
/ December, Eds. Tan, T.S. et al.
Published by Taylor and Francis Group,
London, Vol. 3, pp 2003 - 2045.
24. Menkiti, C. O., Long, M., Kovacevic, N,
Edmonds, H., Milligan, G. W. E. and
Potts, D. M. (2004). “Trial excavation for
cut and cover tunnel construction in
glacial till- a case study from Dublin”.
Skempton Memorial Conference,
Imperial College London, March,
Advances in Geotechnical Engineering,
Eds. Jardine et al., Published by Thomas
Telford, Vol. 2, pp 1090 - 1104.
25. Skempton, A.W. (1954). “The pore
pressure coefficients A and B”.
Géotechnique, 4, No. 4, pp 153 - 173.
26. Skipper, J., Follett, B., Menkiti, C., Long,
M, Clarke – Hughes, J. (2005). “The
engineering geology and characterisation
of Dublin Boulder Clay”. Quarterly
Journal of Engineering Geology and
Hydrogeology (QJEGH), 38, pp 171 –
187, August.
h Soft H s
Stiff
Table 1. Summery of Case Histories - Propped Walls in Dublin Glacial Deposits
Case Location Soil at Soil strength H (m) h (m) Support s* (m) Wall type EI (kN/m2) Del. h (mm) Del. v (mm) Referencehistory dredge lev. su (kPa) configuration
26 Jervis St. Shopping Cen. DBC SPT N = 50+ 9.7 3 Single prop 8.5 Secant 1254800 3 0 Dougan et al. (1996)27 Clarendon St. carpark DBC N = 50+ 6.2 1 Single prop 5 Soldier pile 3895 7 0 Long (1997)28 M&S Grafton St. DBC N = 50+ 7.2 3 Single prop 6 Sheet piles 58500 5 2 Long (1997)39 Dáil Eireann DBC N = 50 to 100 6 0 Single prop 6 Secant 381700 3 0 Brangan (2007)40 Hilton Hotel, College St. Gravel over wth rock N = 50+ 6.3 4 Single prop 6.3 Secant 1254800 1.2 0 Long (2002a)41 ESAT, Grand Canal DBC 4 2 Single anchor 3.5 Secant 347000 5 ?? UCD files42 TCD - Lecky Gravel over DBC N = 30 to 100 7.2 0 3 anchors 2.5 Secant 347000 5 ?? Brangan (2007)43 TCD - Nassau Gravel over DBC N = 30 to 100 8.6 0 2 anchors 3.5 Secant 347000 7 ?? Brangan (2007)44 Ely Place DBC N = 47 to 100 3.5 0 Single prop / anchor 3.5 Secant 201300 3 ?? Brangan (2007)45 Harcourt St. Gravel over DBC N = 28 to 100 4.5 0 Single prop 3 Secant 381700 4.7 ?? Brangan (2007)46 Kings Inn St Gravel over DBC N = 30 - 45 5.7 3.9 Single prop 5.7 Secant 381700 5.5 ?? Brangan (2007)47 Balbriggan Gravel over stiff clay Med dense / stiff 5.7 2.5 Single anchor 5.7 Secant 644126 9 ?? BLP files36 Clancy Barracks Gravel over DBC Dense / hard 5.4 1.5 Single anchor 5.4 Secant 644126 1.5 ?? BLP files in progress48 Westgate Gravel over DBC N = 20 to 100 14 2 Single anchor 14 Secant 644126 7 ?? BLP files49 Tallaght Centre DBC N = 20 to 100 11 1.5 Single anchor 6 Contiguous 293620 18 ?? BLP files50 TCD - Sports Centre Gravel over DBC N = 30 to 100 6.9 1 2 to 3 anchors 3.5 Secant 644126 9.5 ?? BLP files51 DPT Northern C&C - DP8 DBC N = 30 to 100 12 1 Single prop 10.2 Diaphragm 4320000 4.5 ?? Curtis and Doran (2003)52 DPT Northern C&C - DP36 DBC N = 30 to 100 17 1 Two props 5.5 Diaphragm 4320000 8.5 ?? Curtis and Doran (2003)53 DPT - Shaft WA2 DBC N = 30 to 100 25 1 Single ring beam 12 Diaphragm 8437500 8.5 ?? Cabarkapa et al. (2003)
h Soft H s
Stiff
Table 2. Summery of Case Histories - Cantilever Walls
Case Location Soil at Soil strength H (m) h (m) s* (m) Wall type EI (kN/m2) Del. h (mm) Del. v (mm) Referencehistory dredge lev. su (kPa)
Stiff soil at dredge
1 Benwell Rd. London clay Stiff 4.5 ? 6.3 Contiguous 100661 11 ? Fernie and Sukling (1996)2 A329-Reading London clay 80 (UU) 6.9 1 9.66 Diaphragm 2500000 18 ? Carder and Symons (1989)3 Leith House London clay Stiff 3.7 6.6 5.18 Contiguous 465950 0.7 ? Thompson (1991)4 Sanct Bld London clay Stiff 5.3 2 7.42 Diaphragm 1280000 4 ? Thompson (1991)5 Newport Crt London clay 180 (UU) 3.7 4.8 5.18 Diaphragm 1280000 8 ? Wood and Perrin (1984)6 Putney Centre London clay Stiff 4.8 0 6.72 Diaphragm 1280000 3.3 ? Thompson (1991)7 British Library London clay Stiff 4 3 5.6 Secant 2571750 20* 20 Raison (1985)8 Bell Common London clay 130 (UU) 5.2 4 7.28 Secant 2330250 10 18 Tedd et al (1984)9 Dunton Green London clay 70 (UU) 8 0 11.2 Contiguous 4385400 38 ? Garrett and Barnes (1984)
10 Broadgate 5 London clay 200 (UU) 9 6 12.6 Contiguous 465950 8 ? Ove Arup & Ptns Files11 Broadgate 9/10 London clay 200(UU) 7.2 6 10.08 Contiguous 465950 10 ? Ove Arup & Ptns Files12 Nat Gal Ext London Clay Stiff 2.3 4.2 3.22 Secant 618000 3 ? Long (1989)13 Bentalls KuT London clay Stiff 6.5 2 9.1 Secant 2855000 12 ? Sherwood et al (1989)14 Waitrose KuT London clay Stiff 3.5 2 4.9 Contiguous 1132800 8 ? Ground Engineering (1985)
15 Swindon Kimmeridge Stiff 4.5 ? 6.3 Contiguous 179600 12 ? Fernie and Sukling (1996)16 Channel Tunnel Gault 60 (UU) 2.7 0 3.78 Sheet 73500 11* ? Young and Ho (1994)17 Cheltenham Lias 120 (UU) 10.6 0 14.84 Contiguous 4141760 40* ? Ford et al (1991)18 Batheaston Lias Very stiff /hard 8.5 0 11.9 Diaphragm 1280000 16 ? La Masurier (1997)19 Finchley Boulder clay 200 (UU) 5.2 1 7.28 Contiguous 1198500 11 0 Brookes & Carder (1996)
20 Manchester Sandstone ? 5 ? 7 Contiguous 254250 11 ? Fernie and Sukling (1996)21 Manchester Sandstone ? 5 ? 7 Contiguous 254250 11 ? Fernie and Sukling (1996)22 Edinburgh Coal Meas ? 6 ? 8.4 Contiguous 245400 6 ? Fernie and Sukling (1996)
23 Salzburg Gravel, clay ? 11.5 ? 16.1 Soldier Piles 1280000 100 30 Breymann (1992)24 Konstanz Gravel, clay ? 7.2 ? 10.08 Secant 639000 26 29 Goldscheider and Gudehus (1988)28 Bloomington, Minn Fill Stiif 7.9 ? 11.06 RC 1055000 4.5 ? Bentler and Labuz (2006)
29 Ship St., Dublin DBC, grav V. stiff / dense 3 2 4.2 Secant 95426 2 ? UCD files30 Intel, Leixlip Till V. stiff / dense 6.8 0 9.52 Contiguous 27600 2 ? Long (1997)31 Schoolhouse Lane DBC V. stiff / dense 5.5 0 7.7 Contiguous 190850 2 ? Long (1997)32 Hammond Lane Gravels Dense 4 2 5.6 Secant 381700 1.5 ? Brangan (2007)33 Parnell St DBC, grav V. stiff / dense 4 2 5.6 Contiguous 190850 2.5 ? BLP Files34 New St. DBC, lmst V. stiff / dense 5.5 2 7.7 Secant 282740 11 ? BLP Files35 North Wall Quay Complex Estuarine 7.5 2 10.5 Secant 644125 2 ? BLP files in progress36 Clancy Barracks DBC, grav V. stiff / dense 7 2 9.8 Secant 260870 1 ? BLP files in progress37 Cork St DBC V. stiff 6 2.5 8.4 Contiguous 238550 7 ? BLP Files38 Ballycullen Rd. DBC V.stiff 7 0 9.8 Contiguous 282740 18 ? BLP Files
51/52 DPT Northern C&C DBC V. stiff / hard 7 1 9.8 Diaphragm 4320000 3.2 ? Curtis and Doran (2003)
Soft soil at dredge
25 UOB Singapore Soft clay 30 (vane) 3 32 4.2 Diaphragm 4320000 10* Wallace et al (1992)26 San Francisco Soft clay soft 10 ? 23.5 Sheet ? 61000 220* ? Clough ??27 Thessaloniki Soft clay 30-40 (vane) 6.5 11.5 9.1 Diaphragm 14000 62* ? Hadjigogos and Avdelas (1998)39 Thorncastle Soft clay SPT 10 4 13 5.6 Secant 381700 23.5 ? Long et al. (2002)
* s = H or H + fixity
